OBJECTIVES MAIN OBJECTIVE
The aim of the REACHOUT study is to maximize the equity, effectiveness and efficiency of CTC health care services in rural areas and urban slums. The qualitative study is part of a context analysis whose purpose is to develop an analytical framework that will be used to design improvement cycles and to explore barriers and facilitators, opportunities and constraints in existing CTC programmes in Mozambique.
SPECIFIC OBJECTIVES
The objectives defined for the context analysis were to:
identify evidence of interventions that result in effective, efficient and equitable care by community health care providers;
map the types of community health care providers;
evaluate the structures and policies of the health system to identify the strengths and weaknesses in the organization of community health services and their management;
identify and assess contextual factors and conditions that form barriers to and facilitators for the performance of community health care providers; and synthesize evidence about the main barriers and facilitators to be developed in
future interventions and identify knowledge gaps to be filled in relation to community health services.
STUDY DESIGN
This exploratory qualitative study followed a participatory approach to the discussion of APE-related work in the study communities. Qualitative research was conducted through in-depth interviews (IDIs) and focus group discussions (FGDs) to explore and understand all aspects regarding the development of the improvement cycles. This approach allowed a full and detailed identification of issues relating to experience and context as well as regarding the APEs’ activities and programme. IDI and FGD topic guides focused on context and community, programme management and the experiences and perceptions related to the APE programme. IDIs were conducted with APEs, supervisors and community leaders, and FGDs with APE clients (mothers of children under five years of age).
22| P a g e DESCRIPTION OF RESEARCH SITES/DISTRICTS AND ADMINISTRATIVE POSTS
The fieldwork was conducted in two districts: Manhiça and Moamba. The selection criteria for these districts were based on their geographical situation, being hard to access and having the APE programme running (in line with the REACHOUT principles regarding the quality and equity of community health services).
Manhiça
The district of Manhiça is located approximately 80km from the northern part of Maputo Province and has a population of about 192,638 inhabitants; more than 50% are young and female. Its health network, despite being significantly evolved, is still insufficient to meet the national standards for health service provision; instead, there is one health facility for every 8730 inhabitants, one bed for every 665 people and one health care professional for every 1600 people (MAE, 2005a).
The epidemiological status of Manhiça district and community is dominated by malaria, diarrhoeal diseases, sexually transmitted infections and AIDS, which represent almost all notified cases every year. It is estimated that 58% of the population is illiterate, with differences between women and men in the administrative posts visited: in Calanga 70.2% of women and 46.4% men are illiterate; in Maluana 63.4% of women and 37.9% of men (MAE, 2005a).
Moamba
The district of Moamba is located in the northern part of Maputo Province and has a population of about 43,396 inhabitants, mostly young and female (53%). Its health network, despite some progress, is still insufficient to improve on the following average indices: one health facility for every 7800 inhabitants, one bed for every 419 people and one health care professional for every 1390 people (MAE, 2005b).
The epidemiological status of the district is dominated by malaria, pneumonia, HIV and sexually transmitted infections (MAE, 2005b). The similarities between Manhiça and Moamba, especially in the communities visited, include the difficult access to health services due to the poor coverage of the health network and deficient access to roads and transportation. Factors related to transportation, the poor coverage of the health service network, drought and illiteracy mean that these communities are vulnerable with regard to inequitable access to quality health services.
Administrative posts
Within the districts, administrative posts were chosen by convenience sampling, taking into account geographical location, the number of APEs, access to health facilities, the presence of older APEs and geographical accessibility. According to these criteria, two
23| P a g e
administrative posts were selected in Manhiça (Maluana and Calanga) and one in Moamba (Sabié).
STUDY POPULATION AND RECRUITMENT STRATEGY FOR STUDY PARTICIPANTS
In both districts (Manhiça and Moamba) and their selected administrative posts, the APEs were selected based on representation in terms of age, sex and geographical location. Mothers (service clients), supervisors and community leaders were chosen, taking into account the community where the APEs worked. The managers included in the study were selected based on being the representative at district level overseeing the APEs. The term gestores(‘managers’ in Portuguese) includes APE supervisors at health facility and district levels, and we have analysed these together as one category of managers.
The REACHOUT project team held sensitization meetings with participants and personnel interested in the project, to explain the project and their participation. These were national, provincial and district-level directorates of the APEs, health facility supervisors of APEs and community members. Almost 15 days before the study was carried out, the team sent a formal letter to the province- and district-level directorates and requested permission for the study. During the study, the district coordinators of the APEs supported the team in locating the selected APEs, and support was also given by the community leaders.
The selection of health facilities and communities was done in collaboration with the District Health Directorates of Women and Social Affairs, taking into account the availability and provision of the APEs and the access roads to these communities.
All participants were informed about the study by the research team, and were asked whether they wanted to participate. People who agreed to participate were requested to give individual consent.
DATA COLLECTION INST RUMENTS
Topic guides were developed for use in all IDIs and FGDs to ensure that all issues were effectively covered. The use of semi-structured topic guides allowed the respondents themselves to dictate the flow of discussions with guidance from the moderator, and a qualitative approach is more flexible than a quantitative approach.
Four data collection tools (one FGD and three IDI guidelines) were developed (see Annex 3):
topic guide for IDIs with APEs;
topic guide for IDIs with supervisors/managers; topic guide for IDIs with community leaders; and
24| P a g e
topic guide for FGDs with mothers.
DATA COLLECTION PROCESS, DATA PROCESSING AND DATA ANALYSIS
The fieldwork took place during July and August 2013 in Maputo Province, in the districts of Manhiça and Moamba. The research team collected data under the supervision of the Principal Investigator.
During the fieldwork, the district supervisors supported the research team on the location of the communities and the first contacts with the APEs. The APEs supported the team in communicating and selecting participants for FGDs and IDIs with community leaders and mothers of children under five. Guided by these procedures, 18 communities were visited. Ten were from Manhiça — namely, Chichongue, Barrica, Doane, Lagoa Pati, Pateque, Pondzoene, Maluana, Chirindza, Mobane and Calanga; and eight were from Moamba: Mahungo, Goane 2, Mukhakazi, Sabié-Valha, Sabié- Missão, Baptine, Langa-Boi and Mavunguane.
Community leaders, clients and APEs were interviewed in their own communities, with the exception of some in Moamba who were interviewed in their district headquarters during a continuous training. Health managers were interviewed at their workplaces, district directorates and health facilities. The interviews were recorded by digital voice recorders and kept in secure files belonging to the Mozambique REACHOUT team.
Portuguese transcripts were made from digital recordings (that were in Portuguese or local language depending on the circumstances and the ability of respondents to ask Portuguese) and double-checked by another researcher. The qualitative data analysis was performed by reading and re-reading the transcripts and identifying emerging themes and sub-themes. A frame for coding (see Annex 5) was developed based on the draft REACHOUT framework, used to generate the topic guides, and themes arising from the data/transcripts. Transcripts were entered and coded with management software used for electronic qualitative data analysis (Nvivo 10). Then, queries were run according to the main codes and sub-codes, and more complex queries looking at sub-groups. Draft narratives were written, reviewed and discussed. Additional analysis was performed to identify the contextual factors that need to be taken into account for the development of the first cycle of improvement.
QUALITY ASSURANCE/TRUSTWORTHINESS
Care was taken in the following ways to ensure that the data collected were accurate: the researchers were familiar with the data collection tools and were given
clear guidance by the Principal Investigator;
only researchers with experience in data collection and committed to data quality were used for the study;
25| P a g e
the IDIs and FGDs were recorded, transcribed and verified independently by researchers and the Principal Investigator; and
a group of local experts and partners of the Liverpool School of Tropical Medicine (LSTM) and KIT with extensive experience in social sciences, health and gender were involved in the analysis to ensure that the data were interpreted from a variety of professional backgrounds. Stakeholders brought different perspectives to the study and data.
Due the respondents’ limitations in communicating in Portuguese, including most clients, community leaders and some APEs, we had to use local languages during almost all interviews and FGD discussions, as a way to guarantee the quality of data collection.
The fieldwork was carefully monitored and continuously supported by the Principal Investigator. Each researcher presented daily progress of the activities to the Principal Investigator, ensuring that all work was carried out as scheduled.
STUDY LIMITATIONS
As related to our approach, the recruitment of study participants is considered a limitation, as APEs were involved in selecting and recruiting participants for the community FGDs with mothers and the IDIs with community leaders; this may have led to bias in these respondents’ answers. Another issue of bias in sampling may have been that APEs tended to select participants who were living in close proximity to where the discussion was to be held (the village with the health post to which the APE was attached) and, therefore, also lived closer to the health post and had easier access to the APEs and their services. Other APEs mentioned that they might have selected those they knew well or with whom they had a good relationship.
The decision to limit interviews with service users to women with children under five was made in view of the focus on children’s health, and this presents a possible limitation given the findings that many APEs focus on curative services and often treat adults as well.
Many rural areas in Mozambique are hard to access by cars (including 4x4 vehicles), bicycles or even walking, and this meant that time management of data collection became a concern; this was another reason to select respondents not living too far from easily accessible locations.
Furthermore, as is the case of many developing countries, in Mozambique the level of education of community members, including our respondents, is relatively low, and this influenced the level of understanding during data collection as well as the need for
26| P a g e
additional layers of translation, from local languages to Portuguese and from Portuguese to English, with a potential loss of fidelity in transcribing.
ETHICAL CLEARANCE
Ethical clearance was obtained from the ethics committee at KIT in the Netherlands and the Institutional Bioethics Committee of UEM. Administrative approval was obtained from the Maputo Province Health Directorate and the District Health Directorates of Manhiça and Moamba. The study implementation adhered to good research practices. For example, the purpose and objectives of this study were thoroughly explained to all potential participants, and only those who agreed were enrolled in the study. Potential participants were informed that participation was voluntary, that there were no harms or other negative consequences for those who declined to take part in the study, and that all information obtained during the study would be anonymized and stored with strict adherence to confidentiality norms.
27| P a g e